@assistant motion vote 6f361693a7b248730b41d4292f89dc6f6f166bc8
Hi @cryptog
Here are the details for the Motion Vote on 6f361693a7b248730b41d4292f89dc6f6f166bc8:
##[6f361693a7b248730b41d4292f89dc6f6f166bc8][1]
[1]: https://discuss.nubits.com/t/voting-motion-to-make-the-nu-source-code-available
Blocks: 2318 (23.180000%
)
Share Days: 731653176 (23.806538%
)
This is the finalize motion to make the Nu source code open to the public. For more information about this proposal, and to view discussions about what items some community members belived should be completed on the development roadmap before they will vote for this motion, see:
The motion is stalling at approximately 23-24% support. Perhaps now would be a good time for those voting ānoā to discuss what sorts of timelines and milestones they are waiting for prior to casting their vote? It is important to give outsiders who are waiting to contribute the opportunity to see where community consensus stands.
I am concerned that, with the source code available, it would be easy for a set of clones to pop up, confusing the marketplace with too much choice, and reducing the amount of investment / interest in any one effort. Also, even with good intentions, others may execute poorly and taint the pegged concept.
Second to that, with $75MM recently heading Coinbaseās way, itās clear that there is capital in the ecosystem. Iād much rather that capital go to building infrastructure around the edge of Nu and the crypto ecosystem, rather than by some VC taking a fork of the code and pumping it by marketing. Money makes people act strangely sometimes.
Third to that, Iād like to keep the attention exclusively on Nu for recruiting / retaining custodians. This precious resource may also be made scarce by clones.
Whereas it may take time, JordanLee and the dev team have proven themselves to have solid talent and the organizational discipline to keep moving ahead. And they are funded and can pay developers at a sensible rate so the time in is being rewarded. Iām willing to wait for this idea to be brought to the next level under a system that seems to be working than to try to accelerate it or add more cooks.
For me, the threshold is when I see the Nu ecosystem at ~$100MM+ (practically out of reach of VC money), or a major exchange not taking on Nu because the client is closed source (and in such a case Iād support a motion to sharing the source code under an NDA with them). The risk then crosses over to undiscovered flaws / security issues compromising the network as too great a risk, something that open source could mitigate thru peer review.
To that end, before the code is open-sourced, it would be great to send the code out for an audit. It may not catch everything, but could reduce the risk of a day-0 exploit taking down Nu.
Linus Torvalds worked on Linux on his own for a better part of a year, and it wasnāt for about another year after that did he adopt the GNU license. And there werenāt sharks circling in the waters around him.
This argument was used by proponents of bitcoin to discard any alt.
The danger of open sourcing Nu is giving the basis for someone else to create the next paradigm after Nu as Nu was created based on the source code of bitcoin but i do not think it would happen soonā¦
Here is a list, based on @desreverās but with my status comments:
(randomly order)
-
NuBot : a. multiple-custodians per pair support; b. protection against mirror attacks; c. multiple products support (now only NBT)
More work is required, but expect good progress in short term with new dev joining the team -
Custodians : 10 active LPC;
Iām concerned about the low numbers, but one can argue that open sourcing may lead to more LPC as it creates trust in the system. Willing to let this one go -
Liquidity : 500.000 NBT buy liquidity
This is a concern. Iām desperately waiting for the developments in this area showing the liquidity in other tiers.
This needs to be in order. -
Merchant Adoption : a. merchant directory ; b. merchant tools to accept NBT;
I hope to achieve more merchant adoption, either direct or indirect with Android client and Shapeshift integration soon -
Other products : An inflation-adjusted currency active for at least 1 month
I think I prefer a currency pegged to CNY as inflation adjusted is still too complicated in the short term.
We should develop and explore this a bit further as someone else might just take this and launch CNY pegged coin based on NuBits source code. -
Tools : block-explorer improved
I think this is well underway, but a bit slow recently. Not a showstopper for me though. -
Tools : at least one working Android NBT wallet
Done -
Tools : the reporting tool for custodians
Underway and close I believe. -
Tools : website with stats about the network online
Nice, but not a showstopper to me -
Media : a video explaining NuBits
Done -
Media coverage : a) at least two major crypto news-site article b) at least one major non-crypto news-site article about us
Got at least one, another one could be part of communications on going open source. Non-crypto might be out of reach for now. -
Community : 1000 users registered on discourse
This is still not where I would hope we would be, open sourcing might contribute here though. Not a showstopper. -
Development Process : Design an open source governance model driven by shareholders
Nice, not a showstopper -
Exchanges : NBT traded on 6 exchange (1 major exchange)
Iām reasonably happy with current exchanges. Adding a major one would be good, but not a showstopper -
NuShares : at least 70% of total shares are distributed
Probably going to happen in next weeks⦠-
Nu Client : a) data feeds implemented c) all major bugs known to date fixed c) ⦠(need help here on people aware of the dev roadmap)
Have the datafeeds, always looking for bugs, open sourcing might help here, not a showstopper anymore
In other words from my perspective there are only a few items remaining. I agree with @woodstockmerkle that it will be easy to setup clone with all implications. Competing e.g. with a CNY pegged coin would cut them short in that area. Also the lack of custodians is a worry. We need to find better ways to attract and retain them otherwise others would do that.
Key triggers to me would also be e.g. large exchange or significant potential partner (e.g. LPC) openly stating that they would like to work with us but only when Nu is open sourced. That would outweigh a lot of the above risks.
I would also support a code review before going open-source to reduce the risk of day-0 exploits.
This is the same argument that people had with bitcoin. Theyāve done just fine. Peercoin has been open source from the start, and innovated POS. Itās been the top POS coin until NXT passed it, and even right now the market cap isnāt that different. I think this clone argument is tiring because the history shows that the innovators stay on top if they keep project momentum going.
If someone clones Nu and stays closed source then theyāre in the same boat as us with developers losing interest, or not wanting to support a closed source project⦠If they go open source with their idea (and itās good) we can merge it into our system and continue on. Iām really not worried about NuLite or NuDoge or NuNXT or whatever pops up.
VCās arenāt going to take an interest in Nu. At least not in the codebase itself. Relating the Coinbase VC funding (a centralized company that plays the game within regulations, and has proven itself for almost two years) to a codebase that requires a lot of overhead to maintain in a decentralized fashion is not an apples to apples comparison. I would rather see developers build cool projects that VCās want to invest in that work within the NuBits ecosystem. They canāt do that until weāre open source. Thereās more risk than reward in trusting a closed source network.
At this point I wouldnāt mind seeing how a competitor attracts custodians. Whatever weāre doing isnāt working. Itās already scarce.
We donāt just need more developers working on the core project. We need developers to take all the cool projects that have been coded for crypto over the past five years and port them over to NuBits, or develop unique projects that would only be possible with a stable coin like NuBits. Open sourcing is not just about development of the core wallet or protocol itself. It allows other projects and services to form around Nu that support it.
This has already happened. When I find the tweet iāll post it here but a major exchange with deep USD liquidity will not touch NuBits until weāre open source. (bittrex or bitfinex or someting) Theyāre literally waiting for us to go open source to add us. Theyāre not going to take a leap of faith on us to help us grow. We, as a project and community, have to take that leap. I monitor twitter constantly and see people complaining about not being interested because weāre closed source. I have messages in my reddit inbox from developers upset because they want to develop things for Nu but they canāt because weāre closed source. Those are only the vocal individuals and many more probably just keep walking by because we have a huge lock and chain wrapped around our door. Individual motions for certain individuals to access the source code doesnāt solve that.
This is exactly why would should go open source sooner than later. @sigmike and @erasmospunk are both very talented but flaws happen in code. We should allow MANY people to review the code. Especially changes before currency burning are implemented. The more complexity we add to the codebase before it becomes open the more potential flaws can be exploited upon release. To me, the longer we wait to release the code the greater the risk for exploits. At this early stage we would be able to fix the problems and recover quickly without many people being affected. Why everyone keeps listing that big exchanges, or tons of investment money should be involved with Nu before we even have a proper code audit seems totally backwards to me. People have to trust you before theyāre going to throw money in your direction. Open sourcing now means we can strengthen the code and improve trust rather than potentially break that trust later if a critical flaw is discovered that impacts many people/businesses.
I think we can get a far better code audit for free just by going open source anyway. I understand the concern and desire for this though.
I think he had different intentions than us when he started that project, and he was working from scratch. Weāre a fully released project. Again, I donāt think this is an apples to apples comparison.
As I said in my first reply, crypto-innovations have been favorable to first movers. If we keep project momentum going weāll be fine. Lack of developer interest or community support is what kills projects. I think we can improve both of those areas significantly if we go open source now.
I feel these limitations to NuBot shouldnāt impact our decision to release the main client code. NuBots focus is on crypto exchanges, but opening up the main codebase would allow for projects like the bitcoin lighthouse to be built. Something like that has nothing to do with NuBot and would help give more uses for NuBits right now. NuBot is important to the project, but I donāt think we should live in this binary mindset that Nu can only exist with NuBot. There may be better ways to approach our operation and giving people access to the code could help spark that path to discovery.
Trust goes a long way. We can greatly increase trust by going open source. We can also prevent catastrophic flaws in the system by squashing them early before too many people would be impacted by going open source much later.
I donāt think offering an inflation-adjusted currency provides any benefit to the network, and others on the team have shared this sentiment. Lets focus on doing one think really well right now, and cleanse the codebase through open audits before it gets too complex with all these other currency and asset ideas.
Many more block explorers can be created if people had access to the code.
(we have two, and one supports NSR)
This is such a weird and arbitrary point. Considering many of the top crypto-news networks are pay to play I donāt think we should let their coverage impact the decision making process of our projects milestones.
Again. Such an arbitrary number. I donāt think this should even be a point of consideration to be honest.
Major exchanges arenāt going to trust binaries they didnāt compile themselves. Weāve already had one exchange tell us this directly on twitter. They have more to lose than they have to gain. Itās in our best interest to make them feel comfortable supporting us by releasing the code.
This is the only point that I would agree with, but given that the difficult has increased significantly since JLās last update at 60% distributed I would say weāre at a very comfortable distribution by now. I would be happy with 50%+ but I do understand the desire here.
TL;DR
Nu is ready to drop the diapers and enjoy running around outside on its own. There may be some bumps and bruises attributed with that, but itās part of the growing process. Open audit of the code base is the cheapest way to verify we have a secure product before people have a major reliance on Nu that could be impacted through discovered flaws. Through the trust of having an open system and tightening all the loose ends weāll be able to attract bigger exchanges, more ancillary projects, and in turn allow developers to build a diverse ecosystem of services related to Nu without the core team acting as gatekeepers. Even if the motion passed tomorrow it clearly states that weāll have up to a month and a half to release the code. Thatās ample time to address existing concerns and to prepare for any unknowns that may arise after code release.
I know we all want what is best for the project, but I think many of the existing concerns are either unfounded or have been resolved already. Letās take the next step for NuBits to grow as a crypo-project and stop living in fear. The potential is great for this to be really bad or really good for Nu, but the outcome will not be based on a coin flip. It will be based on the continued effort of the developers and community to move Nu forward. I think we have enough momentum to reach that really good potential and we canāt do it until we make it past this milestone.
I have nothing substantial to add here.
Itās time to show some balls!
ā¦in my opinion it poses a bigger danger to Nu to stay closed source than it does going open source.
Adoption hindering by staying closed source is a fact.
Other competitors will enter this field of pegged crypto currencies. They donāt need to clone Nuās code to be dangerous for Nu.
They could start open source!
Ever thought about that?
Iām voting to open up the source code.
Iāve been voting for this motion for a while. @CoinGameās points echo a lot of my own reasons for it.
Iād like to add to what @CoinGame mentioned that JL already described the idea of Nu very precisely in the whitepaper. So with a Coinbase like investment you could take the whitepaper, hire 20 smart developers and develop a client based on the current Peercoin development tree in a short time.
So the Nu idea itself is āopen sourceā for a long time, its only the very practical implementation weāre talking about here. I think the upcoming competitors to Nu are already in development, even those which aim to use a similar concept. They donāt just sit there and wait for the source code to pop up on github.
But I wouldnāt be keen to speed them up by providing the code they need, so not convinced.
Will add this to one of the triggers I have to start voting in favor of this motion: The imminent launch or launch of a new stable coin competitor.
Edit: In many occasions developers donāt need to wait for open source as they can already work on NuBits applications as we have this opensource library available: https://github.com/Cybnate/NuBitsj
BTW good debate, I guess this one will go into history as the most debated motion for a while.
They donāt need our code. Bitshares didnāt need our code for bitUSD. Whoever created CoinoUSD didnāt need our code. Ethereums (open source) schelling dollar concept doesnāt need our code. http://pebble.io/ doesnāt need our code. bitreserve isnāt waiting to pounce on us until we go open source. I think the notion someone could step up and immediately clone Nu, or quickly create a competitive variant downplays the months of planning and effort it took pre-launch, post launch, and ongoing to meet our current goals successfully. Thereās more incentive for devs to build within our ecosystem than against it. We have a lot of momentum. I think thereās this narrow view that Nu has some magic golden code that allows us to do something nobody else can do. In fact many other people have stable coin projects. It was a novel idea early last year, but now weāre just one of many. I think itās better to allow people to become familiar with the Nu codebase before other interesting competitors join into the fray. The stable coin race is still early, but weāre not running it alone at this point. Itās the same reason Microsoft and other major companies give out software to students for free. So that when they start working their experience is with using Microsoft products and technologies. Opening up our code allows people interested now in the idea of a stable coin to become familiar with us, and not some other shiny new ideas that are probably in development right now.
I think a more accurate statement would be:
On a couple occasions NuBits developers have worked directly with other developers to help them build stuff. On many occasions developers immediately lost interest in the project or didnāt trust it because itās closed source.
The messages Iāve seen regarding upset devs far exceeds those weāve worked with. It still doesnāt expose our code to the light of peer review that would further legitimize NuNet in the eyes of many.
We are digressing, but people donāt need to work with NuBits developers to use this opensource library.
Unfortunately I havenāt been exposed to all those upset devs you have seen messages from, so I have to take your word for it. I think a good number of them might just not know what is already available to them.
And NuBits is not only about code for a stable coin, there is also a voting system and other features not widely available yet, it is the set of tools what makes it pretty unique. A number of features are already open; the bots are open source, key library is open source, just the core code is not yet. And for good reasons as I think. It is unique for now.
And again if more interesting competitors join the scene or projects with similar code/functionality as NuBits then that will be a trigger for me to vote in favor of this motion immediately.
I have a hunch that, even if all forum readers voted for this motion, it might still struggle to pass. Iām guessing there are a lot of unconfigured minters out in the wild, āaccidentallyā voting for 0% interest, 0 custodians, and 0 motions. Itās hard to estimate their impact.
There have been motions / proposals passed very fast recently. I think the slowness here reflects indecision of a lot of shareholders.
Interesting debate here, donāt want to play words but⦠You vote in favor or are reflecting indecision? What about just voting against it?
OK, voting No being impossible, someone start a ānot open source Nuā motion and see how many Yes votes it gets
I think the risk is not in open sourcing the code base. Open sourcing the code base will boost Nuās development for sure.
The risk is in distributing NuShares to people that do not care about Nu.
So we should be very careful about the question of raising money via NuShares sales.
Opening NuShare fully to the public should be gradual but opening the source code should be done quickly in one step.